NNA to Nagios xi Integration

SteveBeauchemin · Post by **SteveBeauchemin** » Thu Jun 08, 2017 2:22 pm

I have had NNA for a long time, but am just now trying to use it from within Nagios xi.

When I drill into a Host and click the NNA tab in Nagios xi, I get a screen like this:

NNA-Spinner.PNG

On the left in the red circle, I get a spinner that just spins forever.
On the right there are 2 hyperlinks. They both show exactly the same data in NNA when I click them.
The link address are both identical. They work fine to open and display the expected data in NNA.

In the Admin Tab, Manage Components, Nagios Network Analyzer Integration settings I have this:

NNA-Spinner-2.PNG

Version 1.2.8 07/28/2016 Nagios Enterprises, LLC

The API is from NNA Admin user. all other users have API disabled in NNA

I do not want NNA to dynamically create hosts in Nagios xi.

I need a little help here getting the screen to populate. What are my next steps?

Thanks

Steve B

Post by **lmiltchev** » Thu Jun 08, 2017 2:30 pm

You are using xi 5.4.2, correct? Do you see any NNA related errors in the apache error log?

Code: Select all

tail -100 /var/log/httpd/error_log

What is the version of NNA that you are currently using?

Also, on which box are you using SSL - xi, NNA or both? I will try to recreate the issue in-house. Thanks!

SteveBeauchemin · Post by **SteveBeauchemin** » Thu Jun 08, 2017 3:34 pm

The web error log is clean - no entries from today. On the NNA system...

The Nagios xi is a different story. I force ssl there so the log file is ssl_error_log. It seems unrelated.

My IP is 151.159.43.53 today for this post.

Every time I click a Host, drilling in from the GUI to see details, I get this line in the ssl_error_log

Code: Select all

[Thu Jun 08 15:02:27.850209 2017] [:error] [pid 36653] [client 151.159.43.53:62700] PHP Warning:  Invalid argument supplied for foreach() in /usr/local/nagiosxi/html/includes/components/xicore/status-object-detail.inc.php on line 2507, referer: https://crulnxicore01.rockwellcollins.com/nagiosxi/includes/components/xicore/status.php?show=services&hostattr=10&hoststatustypes=2&servicestatustypes=28&serviceattr=10

Then I click the NNA tab, and get no more line items in *error_log
I have the latest NNA 2.2.3
Nagios xi 5.4.2

I have them both using SSL. I had previously made sure that both systems were using https. I thought maybe the browser was not allowing insecure data.

More sample data. - bunches of these... 2 line paired message - I'm not sure who is clicking what for these to show up. I think they are something unrelated.

Code: Select all

==> ssl_error_log <==
[Thu Jun 08 15:14:53.830708 2017] [:error] [pid 58899] [client 10.9.52.198:62071] PHP Notice:  Undefined index: id in /usr/local/nagiosxi/html/includes/components/ccm/classes/data_class.php on line 1611, referer: https://crulnxicore01.rockwellcollins.com/nagiosxi/includes/components/ccm/index.php?cmd=modify&type=hostgroup&id=110&page=1&returnUrl=index.php%3Fcmd%3Dview%26type%3Dhostgroup%26page%3D1
[Thu Jun 08 15:14:53.830725 2017] [:error] [pid 58899] [client 10.9.52.198:62071] PHP Notice:  Undefined index: idMaster in /usr/local/nagiosxi/html/includes/components/ccm/classes/data_class.php on line 1612, referer: https://crulnxicore01.rockwellcollins.com/nagiosxi/includes/components/ccm/index.php?cmd=modify&type=hostgroup&id=110&page=1&returnUrl=index.php%3Fcmd%3Dview%26type%3Dhostgroup%26page%3D1

Thanks for the quick response.

Steve B

SteveBeauchemin · Post by **SteveBeauchemin** » Thu Jun 08, 2017 3:40 pm

There are other folks working in Nagios xi adding systems, changing stuff. They had introduced some configuration errors. The BPI was failing, and a commit would not run. After they cleaned up, and got a successful Apply Configuration with a clean run, those errors from the previous post are no longer happening.

I have tail -f *error_log running but it is no longer scrolling with data.

Steve B

Post by **tgriep** » Thu Jun 08, 2017 4:41 pm

Are you still having the issue with the NNA integration in xi?

If so, follow these instruction for editing the php.ini file for both servers.

Edit the /etc/php.ini file and change the following from

Code: Select all

max_execution_time = 30
max_input_time = 60
memory_limit = 128M

to

Code: Select all

max_execution_time = 60
max_input_time = 120
memory_limit = 512M

If they were already increased, increase them more.
Then add this to the bottom of that file

Code: Select all

max_input_vars=5000

Save the file and restart Apache for the changes to take affect.

Code: Select all

service httpd restart

Try that and let us know if it helps.

SteveBeauchemin · Post by **SteveBeauchemin** » Fri Jun 09, 2017 11:00 am

Yes, still having issue with the integration to Nagios xi.

I made the php changes:
NNA Server:

Code: Select all

vi /etc/php.ini

;max_execution_time = 30
; SLB 2017-06-06
max_execution_time = 240

;max_input_time = 60
; SLB 2017-06-06
max_input_time = 600

;memory_limit = 128M
; SLB 2017-06-06
memory_limit = 2048M

; max_input_vars = 1000
; SLB 2017-06-09 - was there but was commented out - changed to 5000
max_input_vars = 5000

systemctl restart httpd

Nagios xi Server:

Code: Select all

vi /etc/php.ini

;max_execution_time = 180
; SLB 2017-06-09
max_execution_time = 240

;max_input_time = 180
; SLB 2017-06-09
max_input_time = 600

;memory_limit = 512M
; SLB 2017-06-09
memory_limit = 1024M

; max_input_vars = 1000
; SLB 2017-06-09 - was there but was commented out - changed to 5000
max_input_vars = 5000

systemctl restart httpd

The Nagios xi GUI behavior is unchanged. If I use the hyperlinks, the page will open on NNA. The spinner still just spins forever.

Steve B

SteveBeauchemin · Post by **SteveBeauchemin** » Fri Jun 09, 2017 11:46 am

I figured out my problem. damn... And I hate that is was just number versus name.

In the Admin component where I define the NNA server, so Nagios xi can reach it. I had used the IP address and not the Host Name.
When I changed it to the host name, the screen came right up in the hosts, and services detail view.

I did modify the component to show a 2 hour view and not a 24 hour view. The 24 hour view takes forever to load.
I have a large site, so it is the price I pay. Everything takes a long time to run. Not just NNA

wow... simple things huh... probably an ssl, http https thing. Was just not obvious.

Last question for now:
There are 2 hyperlinks on the Nagios xi page:

(Open this query in Network Analyzer)

and

See all records and details in Network Analyzer

They are identical. go to the same place. Was that the intent? Was there supposed to be 2 different locations opened when using them.

Thanks

Steve B

Post by **tgriep** » Fri Jun 09, 2017 12:42 pm

It probably was the SSL settings. If the cert of the NNA server were setup to use the hostname and not the IP address, that probably would cause the issue.

About the links, they do the same thing on my system as well.
Was that intended, I doubt it, probably a duplicated line of code in the component that didn't get updated.

SteveBeauchemin · Post by **SteveBeauchemin** » Fri Jun 09, 2017 3:14 pm

Thanks for the help.
The integration of NNA to Nagios xi is awesome. It would be a little better if I could control the default time frame for data display.
I did edit the php and set it to 2 hours.
This can be closed if you like.
Steve B

Nagios Support Forum

NNA to Nagios xi Integration

NNA to Nagios xi Integration

Re: NNA to Nagios xi Integration

Re: NNA to Nagios xi Integration

Re: NNA to Nagios xi Integration

Re: NNA to Nagios xi Integration

Re: NNA to Nagios xi Integration

Re: NNA to Nagios xi Integration

Re: NNA to Nagios xi Integration

Re: NNA to Nagios xi Integration